Deconvolution of adaptive optics images: from theory to practice

摘要

Practical guidelines are proposed for deconvolution of partially adaptive optics corrected images: from raw data to high photometric precision object restoration. preliminary processing is discussed (detector calibration, background substraction...). The deconvolution scheme itself is then presented. It takes into account the noise statistics in the image, the imprecise knowledge of the point spread function (PSF), and the available a priori informaiton on the object (spatial structure, positivity...). This decovolution scheme is first validated on simulated images of NAOS, the AO system of the VLT and then applied to astronomical images. In particular, an edge preserving regularization is tested on several solar system objects: Io, Uranus and Neptune.